Intelligent Energy Management Technologies: ICAEM 2019 (Algorithms for Intelligent Systems)

Key technology Method Result-Advantages Disadvantages Reference 40 B. J. Pandya et al. Component repair Module separation Two kinds of methods to dismantle Can increase the output power of old Can be used only if the external [16] and Recycling of silicon analyze faults in the junction box solar panels junction box and outer layer are aging Organic solvent method Mechanical pressure is critical to Organic liquid waste produced [17] Did not involve the resource recovery [18] Compared to three methods: artificial suppress the swelling of the EVA. The of Silicon [19] Not sufficient for single component [20] disassembly, use of organic solvent to silicon panel was recovered separation; still in the laboratory research stage dissolve components, and heat successfully with no damage Organic liquid waste produced Non-purified silicon wafers treatment The heat treatment process was the Artificial disassembly, crushing, optimal solution cryogenic breaking, and electrostatic Obtained a mixture of different separation types of materials Organic solvent-assisted ultrasonic significantly shortens the dissolution method time of EVA in Heat treatment and chemical etching organic solvents methods Module separation Cement-based thermal insulation Recycling of silicon Organic liquid waste produced [21] system and chemical method (continued)

(continued) 3 Renewable Energy Conversion: Sustainable Energy Development … Key technology Method Result-Advantages Disadvantages Reference Recycling of rare metals Grinding and hydrometallurgy Recycling of indium and gallium High price of chemicals [22] Physical and chemical methods: Recycling of cadmium telluride; net High price of chemicals [23] blasting, cost estimated at Process is complicated [24] mechanical processing and $0.04–$0.06/W All types of mixtures, hard to separate [25] dissolution of the semiconductor, and Recycling (CdTe and CIS) precipitation, plating, and ion Mix of recycled poly-silicon, exchange amorphous silicon, and CdTe Recycling (CdTe and CIS) by wet solar panels; glass directly mechanical recycle treatment such as grinding and flotation, or dry mechanical processing methods such as vacuum blasting Treated multiple types of solar panels together. Two physical methods were used: panels were broken and then heat-treated or broken with a hammer 41

42 B. J. Pandya et al. 5 Conclusion After exhaustive review based on available literature presented by renowned researchers in field of solar power generation using photo volatile technology is discussed briefly here. We have observed various characteristics of solar PV cell and effect of variables in solar PV cells. Generally electrical performance of PV cells and modules are measured by manu- facturer at standard temperature condition but in actual power plant modules operate at different climatic conditions. Hence there is variation in performance, conversion efficiency, and generated power are diverse from experimental results within labo- ratory and real field environment. A noticeable deviation observed for experimental result between solar PV cell and module. Here we have mentioned laboratory-based performance analysis of different PV cells like silicon crystalline, thin-film, organic, etc. and laboratory-based performance analysis of different PV modules; done by various researchers. After comparing it, it is observed that there is less power generated in PV module. In India, poly-crystalline and mono-crystalline PV panels are majorly used so here we have discussed a comparative AHP (Analytical Hierarchy Process) analysis performed by honorable researchers and made summary of it as per different condi- tions. At the end of AHP analysis, we came to the conclusion that mono-crystalline panel of brand P6 is best suitable according to financial, mechanical, environmental, electrical, and customer satisfaction criteria. Here we also discussed novel solar cell technologies and some methods for recycling waste solar panels. References 1. Manishkumar, Arun kumar, Performance assessment and degradation analysis of solar photovoltaictechnologies: a review 2. Sharma A (2011) Comprehensive study of solar power in India and world. India. Renew Sustain Energy Rev 15: 1767–1776 3. Park S, Pandey A, Tyagi V, Tyagi S (2014) Energy and exergy analysis of typical renewable energy systems. Renew Sustain Energy Rev 30:105–123 4. Maehlum MA (2013) Which solar panel type Is best? Mono vs. polycrystalline vs. thin–film. Energy Inf 5. Tyagi V, Rahim NA, Rahim N, Jeyraj A, Selvaraj L. Progress in solar PV technology: research and achievement. Renew Sustain Energy Rev 6. El Chaar L, El Zein N (2011) Review of photovoltaic technologies. Renew Sustain Energy Rev 15:2165–2175 7. Akinyele D, Rayudu R, Nair N (2015) Global progress in photovoltaic technologies and the scenario of development of solar panel plant and module performance estimation− Application in Nigeria. Renew Sustain Energy Rev 48:112–139 8. Gee JM, Schubert WK, Basore PA (1993) Emitter wrap-through solar cell. In: Photovoltaic specialists conference, conference record of the twenty third IEEE, pp 265–270. IEEE

3 Renewable Energy Conversion: Sustainable Energy Development … 43 9. Pandey A, Tyagi V, Jeyraj A, Selvaraj L, Rahim N, Tyagi S (2016a) Recent advances insolar photovoltaic systems for emerging trends and advanced applications. Renew Sustain Energy Rev 53:859–884 10. Goetzberger A, Hebling C, Schock H-W (2003) Photovoltaic materials, history, status and outlook. Mater Sci Eng: R: Rep 40:1–46 11. Green MA, Emery K, Hishikawa Y, Warta W, Dunlop ED (2016) Solar cell efficiency tables (version 47). Progress Photovolt: Res Appl 24 12. The selection of the best solar panel for the photovoltaic systemdesign by using AHP FigenBaloa, LütfüS¸ ag˘bans¸uab, * Energy Procedia 100 (2016) 50–53 13. Pandey A, Tyagi V, Jeyraj A, Selvaraj L, Rahim N, Tyagi S (2016b) Recent advances in solar photovoltaic systems for emerging trends and advanced applications. Renew Sustain Energy Rev 53:859–884 14. Jha AR (2008) MEMS and nanotechnology-based sensors and devices for communications, medical and aerospace applications. CRC Press 15. Xu Y, Li J, Tan Q, Lauren Peters A (2018) Global status of recycling waste solar panels: a review. ChinaWaste Manag 16. Lin W, Chen E, Sun YL (2011) Analysis of old photovoltaic component junction box disassembling mode. Sol Energy 7:26–29 17. Doi T, Tsuda I, Unagida H, Murata A, Sakuta K, Kurokawa K (2001) Experimental study on PV module recycling with organic solvent method. Sol Energy Mater Sol Panels 67:397–403 18. Dong L (2009) Research on waste crystalline silicon solar panels resource recovery 19. Kim Y, Lee J (2012) Dissolution of ethylene vinyl acetate in crystalline silicon PV modules using ultrasonic irradiation and organic solvent. Sol Energy Mater Sol Panels 98:317–322 20. Klugmann-Radziemska E, Ostrowski P (2009) Chemical treatment of crystalline silicon solar panels as a method of recovering pure silicon from photovoltaic modules. Renew Energy 2009:1–9 21. Fernandez LJ, Ferrer R, Aponte DF, Fernandez P (2011) Recycling silicon solar cell waste in cement-based systems. Sol Energ Mat Sol C 95:1701–1706 22. Guangdong Xiandao Rare Material Co. Ltd (2011) Recovery of copper indium gallium selenide thin-film solar panel, involves crushing solar panel, soaking in sulfuric acid, filtering, extracting, separating, stripping extraction liquid, adding reducing agent to raffinate and filtering (Chn) CN 103184338-A[P] 2011-12-29 23. Berger W, Simon FG, Weimann K, Alsema EA (2010) A novel approach for the recycling of thin film photovoltaic modules. Resour Conserv Recy 54:711–718 24. Granataa G, Pagnanelli F, Moscardini E, Havlik T, Toro L (2014) Recycling of photovoltaic panels by physical operations. Sol Energy Mater Sol Panels 123:239–248 25. Kang H, Hong T∗, Jung S, Lee M. Techno-economic performance analysis of the smart solar photovoltaic blinds considering the photovoltaic panel type and the solar tracking method 26. Hafez AZ, Yousef AM, Harag NM (2018) Solar tracking systems: technologies and trackers drive types –A review. Renew Sustain Energy Rev 91:754–782. https://doi.org/10.1016/j.rser. 2018.03.094 27. Heslop S, MacGill I (2014) Comparative analysis of the variability of fixed and tracking photovoltaic systems. Sol Energy 107:351–364. https://doi.org/10.1016/j.solener.2014.05.015 28. König D, Casalenuovo K, Takeda Y, Conibeer G, Guillemoles J, Patterson R et al (2010) Hot carrier solar cells: principles, materials and design. Phys E: Low-Dimens Syst Nanostruct 42:2862–2866

Chapter 4 Evaluation of Thermal Degradation Behavior of Cardboard Waste Samit Kumar Singh and Sadanand A. Namjoshi 1 Introduction The waste-to-energy is really a challenging task and to accomplish this task, the pyrolysis is one of the best options. The composition of MSW varies because of socio- economical status of the society, economic development, demographic division, and waste collection potential also for waste-to-energy options. The extensive rises in the MSW generation having severe impact on environment but its disposal associated with social and economic problems. The pyrolysis having gas, char, and liquid yield products; it means three types of fuels are available. The significance of pyrolysis is also due to the decomposition of polymers into fuels which are not easily degradable and contaminate atmosphere heavily when thrown as a waste at open places. The design of pyrolyzer is based on kinetics studies which correlate the weight degradation with respect to time and temperature; it also represents space or in other words, volume requirement of fuel in the form of pyrolytic products with respect to time temperature. The quantities of yield products are also depends on available space otherwise temperature gradient will exist and conduction losses increase. Garcia et al. [1] focused kinetics of pyrolysis at different heating rate between 1.5 and 200 °C min−l. After observing the collected results, a theoretical study was carried out in connection with the effect of heating rate and the transmission of heat on the kinetic parameters which can be obtained using the experimental data. Fontet al. [2] investigated the kinetics of the almond shells impregnated and non- impregnated with CoCl2, polyethylene lignin, and MSW with the help of an analytical S. K. Singh (B) · S. A. Namjoshi 45 Madhav University, Pindwara (Sirohi) 307026, Rajasthan, India e-mail: [email protected] S. A. Namjoshi e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2021 M. Shorif Uddin et al. (eds.), Intelligent Energy Management Technologies, Algorithms for Intelligent Systems, https://doi.org/10.1007/978-981-15-8820-4_4

46 S. K. Singh and S. A. Namjoshi pyro probe 1000 or pyroprobe 100, thermogravimetric fluidized sand bed reactor was used to understand the thermal decomposition of heterogeneous materials. Wu et al. [3] reported kinetic study of painted printed and writing paper by TGA in nitrogen atmosphere for the temperature range is from 450–900 K at 1, 2, and 5 K/min heating rates with two stages of decomposition. Park et al. [4] suggested a new method for estimation of kinetics constants of low-density polyethylene (LDPE), linear low-density polyethylene (LLDPE), and high-density polyethylene (HDPE), at heating rates of 100 °C/min and 500 °C/min. The numerous analytic methods mentioned were used for comparing with the kinetic parameters estimated from the dynamic method. Sorum et al. [5] investigated in-depth knowledge of pyrolysis process and estima- tion of chemical kinetics important constituents of MSW. TGA estimates the kinetic constants which are executed at 100 °C/min heating rate in N2 gas atmosphere. The degradation of hemicellulose, cellulose, and lignin was shown by three autonomous parallel reactions the cellulosic fraction of MSW. David et al. [6] explained pyrolysis process depending upon the numbers of TGA experimental work on cardboard. The research work includes the determination of kinetics parameters for cardboard. These research works also focus pyrolysis process with the support of numbers of TGA experimental works also compare the results of graphical method with numerical results which are in good level of satisfaction. Molto et al. [7] focused on kinetic studies for the decomposition of used cotton fabrics and for explaining the behavior of decomposition all the runs were performed and have been proposed and tested. In the pyrolysis of the used cotton fabric waste, the model comprises two parallel reactions during decomposition. Miranda et al. [8] studied thermal degradation behavior of textile waste by TGA at different heating rates also decomposition was studied by semi-batch pyrolysis process. The DTG curves show three peaks; two out of which were hemicellulose and cellulose and one might be due to decomposition synthetic polymer. The major research in present work focuses on thermal degradation behavior of cardboard waste at 5, 10, and 15 °C/min heating rate and also ash obtained during decomposition of cardboard analyzed using Scanning Electron Microscope (SEM). 2 Experimental Methodology 2.1 Material The experiments were conducted on cardboard collected from waste packing boxes which was sheared into small pieces of 1–2 mm by using scissors. The size of the particle used in TGA having micro-weight and sample amount limitation of container. Table 1 indicates proximate, ultimate result, and HHV of cardboard.

4 Evaluation of Thermal Degradation Behavior of Cardboard Waste 47 Table 1 Proximate analysis, ultimate analysis, and higher heating value cardboard Fixed Volatile Ash Carbon Hydrogen Nitrogen Sulfur Oxygen HHV MJ/kg carbon matter (% (% wt) (% wt) (% wt) (% wt) (% wt) 15.35 (% wt) (% wt) wt) 12.6 74.9 12.4 42.13 5.3 0.18 0.54 39.51 2.2 Thermogravimetric Analysis (TGA) The engineering design of modern pyrolysis reactors and gasifiers requires a deep knowledge of kinetics studies for different MSW constituents at different temperature ranges and heating rates. The TGA and DTG were performed in TG/DTG system. Samples’ temperature range varies from room temperature (RT) to 600 °C with 5, 10, and 15 °C/min heating rate in nitrogen atmosphere and flow rate of gas was 50 ml/min. The sample size of cardboard was taken as 6 mg. Samples were placed in platinum container and reference used was α alumina. 3 Kinetics It is necessary to understand the reaction mechanism of pyrolysis for obtaining the reactions with respect to its kinetic parameters. The heated deterioration of MSW constituents is explained to be proceeding through many complicated reactions [5, 6, 8]. A method was formulated to calculate activation energy from single dynamic TGA data employing following model [10, 11]. Here F062 is heating rate, E is activation energy A is pre-exponential or frequency factor, R is universal gas constant (8.314 kJ mol−1 K−1). This equation has no exact solution; hence, different approximations have been done by Flynn wall Ozawa method [12]. For cardboard decomposition, in case TGA, the Flynn wall Ozawa [12] method used for activation energy for three heating rates are used. 4 Results and Discussion Figure 1 represents Thermogravimetric curves of cardboard at 5, 10, and 15 0C/min heating rates, respectively, while Fig. 2 represents the corresponding Differential Thermogravimetric curves for same heating rates. The decomposition is observed in three stages as shown in Figs. 1 and 2. In first stage, moisture present in sample gets removed and hemicellulose gets decomposed. The degradation rate in first stage is slow due to low temperature range and mass

TG (%wt)48 S. K. Singh and S. A. Namjoshi 100.00 80.00 5 C/min 60.00 10 C/min 40.00 15 C/min 20.00 0.00 100 200 300 400 500 600 0 Temperature (°C) Fig. 1 TG curves 500mg/min 5 C/min 450 10 C/min 400 15 C/min 350 300 100 200 300 400 500 600 250 Temperature (°C) 200 150 100 50 0 0 Fig. 2 DTG curves transfer resistance [1, 5, 6, 8]. The degradation of cellulose and lignin occurs in second and third stage, respectively. The similar results are reported by Sorum et al. [5], David et al. [6], Miranda et al. [8] and which substantiates the present work. Figure 1 represents same TG curves pattern for heating rates 5, 10, and 15 °C/min; however, rates and kinetics parameters are differing. During second stage, degrada- tion occurred is faster because cellulose gets decomposed compared to first and third stage where hemicellulose and lignin are decomposed, respectively, as heating rate increases which is represented by TG curves in Fig. 1 [5]. The loss in weight escalates in all three phases at a quicker amount as heating rate rises which is quite obvious that heating rate enhances the values of kinetics parameters too. In connection to three stages of decomposition on the TG curves correspondingly, three peaks are noticed on the DTG curves. DTG curves show as heating rates expand there will be a lateral shift toward greater temperatures and which may be the united effects of the heat transfer at various heating rates and resulting in delayed the decomposition as stated

4 Evaluation of Thermal Degradation Behavior of Cardboard Waste 49 by Sorum et al. [5]. Figure 1 reports that rise in heating rate gives the greater value of activation energy, which can be attributed to the formation of active molecules also enhances in molecular collision at high heating rates [5, 8]. It also shows that the activation energy value is maximum in second stage in comparison to other two stages in all cases all heating rates. During third phase of decomposition, activa- tion energy value gets lowered because of low deterioration rate. It is observed that amount of decaying rises with the rise in heating rate and a declination is observed in total decaying. It is because of the inadequate decaying time for concluding reaction with rising heating rates [8]. In case of all cellulosic fractions, the major weight loss occurs between the temperatures ranges of 250–400 °C as also reported by Sorum et al. [5]. Figure 2 represents the DTG curves which are nothing but decomposition rate with respect to temperature in association with 5, 10, and 15 °C/min heating rates. Results show that rise in heating rate leads to rise in weight loss and corresponding maximum temperature also increases linearly due to delay in decaying which penetrates extra heat into the specimen for shorter while [5]. In case of cardboard, the three peaks are commonly observed and the temperature range in which maximum peak can be observed varies from 175 to 250 °C, 300 to 425 °C, and 450 to 525 °C which corresponding to hemicellulose, cellulose, and lignin, respectively [5, 6]. The description of less marked shoulder on the DTG curves for hemicellulose in case of cardboard is due to the minor content of hemicellulose and/or catalytic effects. Catalytic effects generated by inorganic species such as ash and in the case of cardboard, residues from sulfate process causes the decaying of cellulose to occur at reduced temperature [5]. Tables 2 and 3 present the values of decomposition with temperature range and kinetic constants for different stages of decomposition, respectively. Table 2 % Weight loss in stages of decomposition corresponding to temperature range Heating Stage I Range of Stage II Stage III % Rate temp °C Residual (°C/min) % Loss % Loss Range of % Loss Range of remains in in temp °C in temp °C weight weight weight Cardboard waste 5 6 30–193.5 52.2 193.5–349 27.2 349–545 14.5 10 6 30–193.3 53 193.3–366.5 24 366.5–539.6 17 15 7 30–209.5 50.3 209.5–353.1 27 353.1–532.9 15.7 Table 3 Activation energy Activation energy, E, kJ/mol for cardboard Constituent 1st stage 2nd stage 3rd stage Cardboard 128.15 133.65 83.74

50 S. K. Singh and S. A. Namjoshi Fig. 3 Cardboard char (10 °C/min) The elemental analysis of residues of lignocellulosic waste material, i.e., card- board is performed with the help of SEM using Energy Dispersive Spectroscopy (EDS) technique. EDS is a scientific approach and it employs x-rays that are trans- mitted from the sample when blasted by the electron beam for identifying the elemental composition of the sample. The elemental analysis of residue with the help of scanning electron microscope at 10 °C/min of heating rate has been carried out. The importance of this study is to explore products and elements presence in the residues of cardboard [13]. Figures 3 and 4 indicate elemental analysis of cardboard ash obtained from TGA at 10 °C/min of heating rate. 5 Conclusion For cardboard material, decomposition occurs in three stages. Stage I, II, and III repre- sents decomposition of hemicellulose, cellulose, and lignin, respectively. The esti- mation of decomposition rate is possible through the assurance of kinetic parameters such as activation energy. The activation energy for cardboard waste at the heating rate of 5 °C/min to 15 °C. /min is obtained as 128.15 kJ/mol in first stage, 133.65 kJ/mol in second stage and 83.74 kJ/mol in third stage, respectively. The significance of elemental analysis is to know the possible elements existing in the residue and how they will affect the environment from pollution point of view in later stage.

4 Evaluation of Thermal Degradation Behavior of Cardboard Waste 51 Fig. 4 Elemental spectrum 6 Future Scope of Work The present research work can be extended for development of such reactor at macro-level of cardboard or any MSW constituents for various heating rate makes possible. It also yields products obtained by such reactor which establishes MSW as an alternative better fuel. References 1. Garcia AN, Marcilla A, Font R (1995) Thermogravimetric kinetic study of the pyrolysis of mu- nicipal solid waste. Thermochim Acta 254:277–304 2. Font R, Marcilla A, Garcia AN, Caballero JA, Conesa JA (1995) Comparison between the pyrolysis products obtained from different organic wastes at high temperatures. J Anal Appl Pyrol 32:41–49 3. Wu C-H, Chang C-Y, Lin J-P, Hwang J- Y (1997) Pyrolysis kinetics of paper mixture in municipal solid waste. Fuel 76:115l–1157 4. Park IW, Sea Cheon O, Lee HP, Taik Kim H, Yoo KO (2000) A kinetic analysis of thermal degradation of polymers using a dynamic method. Polym Degrad Stabil 67:535–540 5. Sorum L, Gronli MG, Hustad JE (2001) Pyrolysis characteristics and kinetics of municipal solid waste. Fuel 80:1217–1227 6. David C, Salvador S, Dirion JLM (2007) Determination of a reaction scheme for cardboard thermal degradation using thermal gravimetric analysis. J Anal Appl Pyrolysis 67:307–323

52 S. K. Singh and S. A. Namjoshi 7. Molto J,Font R, Conesa A, Martı´n-Gullo´n I (2005) Thermogravimetric analysis during the decomposition of cotton fabrics in an inert and air environment. J Anal Appl Pyrolysis (Article in Press) 8. Miranda R, Sosa_Blanco C,Bustos-Martı ´nez D, Vasile C (2007) Pyrolysis of textile wastes kinetics and yields. J Anal Appl Pyrolysis 80:489–495 9. Buah WK, Cunliffe AM, Williams PT (2007) Characterization of products from the py- rolysis of MSW. Process Saf Environ Prot 85:450–457 10. Yourlmaz S (2006) Investigation of emissions and combustion kinetics of waste wood samples with thermal and spectral methods. PG Dissertation, Middle East Technical University, Beirut 11. Liu Z (1998) Handbook of thermal analysis. John willy & Sons, pp 47–48 12. Flynn JH, Wall LA (1966) Polym Sci Part B, Polymer Letters 4:323 13. Miskolczi N, Angyal A, Bartha L, Valkai I (2009) Fuels by pyrolysis of waste plastics from agricultural and packaging sectors in a pilot scale reactor. Fuel Process Technol 90:1032–1040

Chapter 5 Wavelength Optimization in Gigabit Passive Optical Network by Proposed Quad Play Architecture Md. Hayder Ali and Mohammad Hanif Ali 1 Introduction Unruly modification in the enterprise resident area network is the prime cause to fulfill the requirements of swift progress of high bandwidth requirements and also changing the nature of enterprise solutions. To maintain a significant solution and to reduce the capEX and opEX, fulfilling the high bandwidth demand, the utmost solution is gigabit passive optical network (GPON) technology [1–2]. ITU-T G.984.x approvals offer GPON system prototypical and does not need any electrical power at the transitional nodes between the aggregation and last-mile device [3]. A big number of researches had been worked on network structures, propagation policies, power consumption budget, bandwidth distribution, and stability of GPON technology [4–6]. The efficiency of service for GPON triple play in IP network and also passive optical network and bandwidth distribution policies are conversed in [7–8]. Ricciardi et al. have exposed an investigation and contrast between EP2P (Ethernet point to point) and GPON solution structure [9]. SDH (E1) traffic with the present triple play explanation is the demand of today’s technological solution. It is much more important to focus on some parameters like output signal range, propagation wavelength, power budget, and sensitivity threshold for convergence solution. In this paper, wavelength is optimized with a proposed quad play architecture by calculating different parameters like path loss, the last end receiving sensitivity, etc. The rest of this paper is ordered as follows. The literature review is presented in Sect. 2. GPON triple play general structure is stated in Sect. 3. In Sect. 4, proposed Md. H. Ali (B) · M. H. Ali 53 Jahangirnagar University, Dhaka, Bangladesh e-mail: [email protected] M. H. Ali e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2021 M. Shorif Uddin et al. (eds.), Intelligent Energy Management Technologies, Algorithms for Intelligent Systems, https://doi.org/10.1007/978-981-15-8820-4_5

54 Md. H. Ali and M. H. Ali quad play architecture is briefly defined. The simulation scenario is in Sect. 5 and performance analysis between GPON triple play and proposed quad play is stated in Sect. 6. Finally, Sect. 7 draws a conclusion to this paper. 2 Literature Review Over the years especially in the last decade, a lot of research study have been passed out to find out the GPON network suitable structure and different type of FTTx model. Some of them are discussed below. Koonen et al. [10] stated fiber-optic technologies and it enables the solution with minimum cost of broadband solution services to the last mile users in multiple-access network architectures [10]. They also designated that the dynamic network recon- figuration can improve system staging. Derek Nesset et al. [11] have investigated GPON structure with a 1300 nm semiconductor optical amplifier and it developed for long spread GPON services [11]. The high gain of −29 dB has empowered a profitable GPON system to function over 60 km and with 128-way split. Claudio Rodrigues et al. [12] have studied and associated the present GPON fiber to the home (FTTH) type service solution, from the point of view of enterprise operator, they also stated that future next-generation passive optical networks (PONs) (XGPONs) and wavelength division multiplexing PONs (WDM-PONs). They also consider some facts like considering into account standardization, wavelength planning, optical line terminal (OLT) as well as optical network terminal (ONT) equipment, and transmission convergence layer [12]. Marcelo Alves Guimaraes et al. [13] have projected a new system for E1 transmis- sion and it allows the fair bandwidth policy, it is performed without circuit emulation techniques by fragmenting E1 signals. Andrej Chu et al. [14] presented an expla- nation by using ant cluster optimization to strategy GPON-FTTH networks with combining equipment [14]. Due to the reflection of many strategy issues such as the number, types, positions of network elements, the optical network development progression frequently parades numerous trials from the optimization point of view. Irfan imitation an estimation of 2.5 Gbps bi-directional GPON based FTTH link using innovative modulation setups [15]. In this research, the presentation of the future scheme was calculated when a single wavelength and two wavelengths were used for triple-play services with different modulation systems. 3 Present Structure of GPON Triple Play In the present GPON triple play structure, the signals enter intoOLT in two separate ways. Firstly, EDFA receives a video signal and pass through. Secondly, data and voice signals enter via a layer-2 switch (ISP connection). These two signals modu- lated into OLT and passed through fiber, optical splitter up to ONT. Finally, end-users receive signals (data, video, and voice) from ONT’s ethernet port (Fig. 1).

5 Wavelength Optimization in Gigabit Passive Optical Network … 55 EDFA LDP ONT Voice BDB ONT Data ISP SW OLT Video Splitter Optical Fiber UTP Cable Coaxial Cable Fig. 1 Present triple play architecture 4 Projected Quad Play Architecture The projected quad play design is not similar as present triple play design. All signals (video and ISP’s data and voice) will enter into a MUX combinedly. These signals will be modulated into OLT. From the OLT, the modulated signal will propagate toward the ONT. The end-users will receive the signals from ONT’s ethernet port as present triple play. The main difference between the present triple play design and projected quad play design is, the modulation done into OLT in the present design and OLT becomes busier and OLT’s CPU uses become high, On the other hand, modulation is done into MUX and that’s why OLT does not disburse time for the modulation (Fig. 2). 5 GPON Quad Play Simulation Environment For better results, the simulation experimented only for single-user connectivity. For this goal, the packet simulation software OptSIM has been used. OptSIM proposes a hierarchical structure of pattern for creating new simulation scenarios. Defined tree levels: • Network model: It’s the primary stage of design. It’s the maximum intellectual and general level. The goals will be to describe network topology, to express network nodes, and to define the statement between each node. • Node model: The second level, objectives are to express the functionalities for each node that is used in the network topology. For all nodes, we built a scheme for scheming inner purposes with the component offered by OptSIM.

56 Md. H. Ali and M. H. Ali Voice Data Video LDP ONT Video Signal OpƟcal Fiber BDB SDH MUX UTP Cable (MUX) ONT ISP OLT SpliƩer Fig. 2 Proposed quad play architecture • Process model: The final level. In this level, a graph is defined for elements used in the second level. The graphs stipulate the jobs performed by the module with the info which will be processed. 5.1 Network Construction To generate the simulation scenario, it is used the tree levels proposed by OptSim. We have agreed to use FTTH topology (no FTTC or others) because the optical fiber reaches straight at the user’s home and it is the resolution that offers more bandwidth capacity. It is considered only this type of control packets in our simulation and ignored others feature related to delays or protocols since the objectives of our analysis need only results about wavelength optimization due to intermediate entree control measures. 5.2 ONU/ONT Design User’s ONT contains splitter, and data earpiece, and video earpieces. Data earpiece is constructed with optical filter, PIN/TIA receiver, and BER Tester. The video signal receiver contains optical filter, PIN/TIA receiver, and electrical filters (Fig. 3).

5 Wavelength Optimization in Gigabit Passive Optical Network … 57 Spectrum Plot-1 Spectrum Plot-2 Video Optical RX Optical Filter-1 Receiver-1 Electrical Multi plot-1 Filter-1 Multi plot-2 Electrical Filter-2 Optical Filter-2 Receiver-2 BER Tester Data/Voice RX Optical Splitter Multi plot-3 Spectrum Plot-3 Optical Filter-2 Receiver-2 BER Tester Multi plot-3 SDH Traffic Receiver Spectrum Plot-3 Fig. 3 Block diagram of ONU/ONT 5.3 OLT Design First, we describe in detail the configurations for Central Office OLT and single end- user ONT and then, will generalize the treatment to all 16 uses. In this experiment, we consider the downstream configuration of GPON with bitrate 1.25 Gb/s and support for quad play. The quad play provision is appreciated as a blend of SDH traffic, data, voice, and video indications. The high-speed Internet element is denoted by a data link with 1.25 Gb/s downstream bandwidth. The voice element can be characterized as VOIP service (voice over IP, packet-switched protocol) and can be collected with data component in physical layer simulations. Finally, the video element can be denoted as an RF video signal (traditional CATV) or as an IPTV signal that also can be mutual with data. In this research, it is considered the former case with an RF video link. To improve the wavelength in PON, the propagation over the optical fiber pathway services the CWDM technique with data/voice component transmitted at wavelengths in the variety of 14,801,500 nm, and video within the 1550–1560 nm range [16] (Fig. 4).

58 Md. H. Ali and M. H. Ali Video Optical TX 1.55 um Signal Plot-1 Signal Plot-2 Sine Wave Spectrum Plot-1 Spectrum Plot-2 Spectrum Plot-3 Generator Summer DML-1 Pre-Amplifire-1 Signal Plot-3 Sine Wave Generator RF Video Signal Optical MUX (55-1000 MHz) BER Tester Electrical DML-2 Pre-Amplifire-2 Generator Data / VOIP TX Signal Plot-4 Spectrum Plot-4 1.25 Gbps Fig. 4 Block diagram of OLT The Central Office Optical Line Terminal block (Transmitter block) involves Data/VOIP and Video elements. The Data/VOIP spreader demonstrated with pseudo- random data originator (PRBS), NRZ modulator driver, direct-modulated laser, and admirer amplifier. The video section demonstrated as RF SCM (subcarrier multi- plexed) link with only two tones (channels) for simplicity. The two channels we used are from standard NTSC analog CATV frequency plan—channel 2 and channel 78 at frequencies 55.25 MHz and 547.25 MHz, respectively. RF video source contains two electrical signal producers, summer, direct-modulated laser, and pre- amplifier. Following, data/voice and video signals are multiplexed at multiplexer and propagated into 20 km fiber span. 6 Result Analysis Applying the loss calculation equation (in Sect. 5), the below table is founded. It shows the last end (end-user) devices (ONT/ONU) signal receiving sensitivity. Minimum receiving sensitivity is the desired one (Table 1). Using 1:32 Splitter with 20 km distance, it is observed that the minimum sensitivity is −34.35 dB at 1550 nm wavelength and the highest sensitivity is −36.66 dB at 1310 nm wavelength (Fig. 5).

Table 1 1:32 Splitter with 20KM distance 5 Wavelength Optimization in Gigabit Passive Optical Network … Wave Attenuation Attenuation Mux Add Drop Demux Filter ODF LDP Splitter BDB ONT Totale Remarks (IN+OUT) Loss (IN+OUT) Loss length (dB/Km) for 10m loss Loss Loss Loss Loss (IN+OUT) Loss Loss (dB) (dB) Loss 1310 0.36 7.2 0.5 1 1 0.5 8.5 0.3+0.3=0.6 0.3+0.3=0.6 17.3 0.3+0.3=0.6 0.3 36.66 1330 0.335 6.7 0.5 1 1 0.5 8.5 0.3+0.3=0.6 0.3+0.3=0.6 17.3 0.3+0.3=0.6 0.3 36.135 1350 0.322 6.44 0.5 1 1 0.5 8.5 0.3+0.3=0.6 0.3+0.3=0.6 17.3 0.3+0.3=0.6 0.3 35.862 1370 0.311 6.22 0.5 1 1 0.5 8.5 0.3+0.3=0.6 0.3+0.3=0.6 17.3 0.3+0.3=0.6 0.3 35.631 1390 0.333 6.66 0.5 1 1 0.5 8.5 0.3+0.3=0.6 0.3+0.3=0.6 17.3 0.3+0.3=0.6 0.3 36.093 1410 0.291 5.82 0.5 1 1 0.5 8.5 0.3+0.3=0.6 0.3+0.3=0.6 17.3 0.3+0.3=0.6 0.3 35.211 1430 0.281 5.62 0.5 1 1 0.5 8.5 0.3+0.3=0.6 0.3+0.3=0.6 17.3 0.3+0.3=0.6 0.3 35.001 1450 0.272 5.44 0.5 1 1 0.5 8.5 0.3+0.3=0.6 0.3+0.3=0.6 17.3 0.3+0.3=0.6 0.3 34.812 1470 0.266 5.32 0.5 1 1 0.5 8.5 0.3+0.3=0.6 0.3+0.3=0.6 17.3 0.3+0.3=0.6 0.3 34.686 1490 0.26 5.2 0.5 1 1 0.5 8.5 0.3+0.3=0.6 0.3+0.3=0.6 17.3 0.3+0.3=0.6 0.3 34.56 1510 0.254 5.08 0.5 1 1 0.5 8.5 0.3+0.3=0.6 0.3+0.3=0.6 17.3 0.3+0.3=0.6 0.3 34.434 1530 0.252 5.04 0.5 1 1 0.5 8.5 0.3+0.3=0.6 0.3+0.3=0.6 17.3 0.3+0.3=0.6 0.3 34.392 1550 0.25 5 0.5 1 1 0.5 8.5 0.3+0.3=0.6 0.3+0.3=0.6 17.3 0.3+0.3=0.6 0.3 34.35 Best 1570 0.25 5 0.5 1 1 0.5 8.5 0.3+0.3=0.6 0.3+0.3=0.6 17.3 0.3+0.3=0.6 0.3 34.351 1590 0.257 5.14 0.5 1 1 0.5 8.5 0.3+0.3=0.6 0.3+0.3=0.6 17.3 0.3+0.3=0.6 0.3 34.497 1610 0.266 5.32 0.5 1 1 0.5 8.5 0.3+0.3=0.6 0.3+0.3=0.6 17.3 0.3+0.3=0.6 0.3 34.686 59

60 Md. H. Ali and M. H. Ali Fig. 5 Receiving sensitivity at different splitting ratio with various distance Analyzing the sensitivity calculation table and graph, it is instituted that 1550 nm wavelength has sensitivity with minimum dB loss. That is, end-user will get better performance. Could use 1550 nm wavelength for GPON quad play instead of using 1310–1610 nm. In general triple play services, the eye diagram could not measure, but in proposed quad play, can easily measure the eye diagram from ONT/ONU’s ethernet port. Figure 6 shows that almost noise-free traffic could be captured from quad play architecture. In Fig. 7, for data connectivity, the blue stripe displays the general triple play sensitivity and the green line displays the projected quad play sensitivity. General triple play has approximately −30 dBm power loss while projected quad play has approximately −25 dBm. In Fig. 8, in the state of video connectivity, the blue line demonstrates the general triple play sensitivity and the green line displays the projected quad play sensitivity.

5 Wavelength Optimization in Gigabit Passive Optical Network … 61 Fig. 6 Eye diagram of SDH traffic (QP) Fig. 7 Data output signal for quad play versus triple play for 1550 nm Fig. 8 Video output signal for quad play versus triple play for 1550 nm

62 Md. H. Ali and M. H. Ali Fig. 9 Total transmitted bit versus number of error The general triple play has approximately −35 dBm power loss while the proposed quad play has approximately −32 dBm. From the above Fig. 9, it is found that for every 1 megabit transmission there is 64-bit error in projected quad play while 76-bit error in a general triple play. If the quantity of transmitted bit increased, then the number of errors will also increase in triple play, but in the projected quad play the error bit will not increase in the same ratio compare to triple play. Form the above Table 2, it can be summarized that if we use 1550 nm wave- length for GPON quad play services, we gain approximately −5 dB for data and voice, approximately −3 dB for video and these are the best gain compare to other wavelengths at GPON triple play service area. 7 Conclusion The main attention was to learn the best wavelength for GPON quad play solution with high performance. Nowadays, the usual triple play facility solution is operating by a varied range of wavelengths. Here, 1310 nm to 1550 nm wavelength is used for triple play connectivity. Sometimes, it makes some intrusion within consumers. It was keen to find out a single wavelength as an alternative of numerous wave- lengths. Analyzing all possible combinations, it is initiated that for quad play archi- tecture, 1550 nm wavelength gives a better eye diagram for SDH traffic and better performance in data, voice, and video service.

5 Wavelength Optimization in Gigabit Passive Optical Network … 63 Table 2 Comparison of different wavelength with dB gain in Quad Play SN Wavelength (nm) Input Output (dBm) Gain Remarks (dBm) (dBm) Best Voice Data Video V = Triple Quad (TP-QP) Video Play Play (TP) (QP) 1 1310 1310 1310 0, −10 −30, − −27, − 3, 0 (V) (V) 26 (V) 26 (V) 2 1490 1490 1490 0, −10 −30, − −30, − 0, 2 (V) (V) 32 (V) 30 (V) 3 1550 1550 1550 0, −10 −30, − −25, − 5, 3 (V) (V) 32 (V) 29 (V) 4 1490 1490 1310 0, −10 −34, − −32, − 2, 6 (V) (V) 29 (V) 35 (V) 5 1550 1550 1310 0, −10 −35, − −30, − 5, 2 (V) (V) 29 (V) 27 (V) 6 1490 1490 1550 0, −10 −35, − −38, − 3, 7 (V) (V) 34 (V) 27 (V) References 1. Cale I, Salihovic A, Ivekovic M (2007) Gigabit passive optical network—GPON. In: Proceedings of the 29th international conference on information technology interfaces, pp 25–28 2. Leo M, Trotta M (2011) Performance evaluation of WDM-PON RSOA based solutions in NGAN scenario. In: Proceedings of the 50th FITCE congress (The Forum for European ICT and Media Professionals), pp 1–4 3. ITU-T Recommendation G.984.1 (2003), 2 (2003), 3 (2004), 5 (2006) and 6 (2007). 4. Frnda J, Voznak M, Fazio P, Rozhon J (2015) Network performance QoS estimation. In: Proceedings of the 38th international conference on telecommunications and signal processing (TSP), pp 1–6 5. Rokkas T (2015) Techno economic analysis of PON architectures for FTTH deployments. In: Proceedings of the conference of telecommunication, media and internet techno-economics (CTTE), pp 1–5 6. Lee SS, Li W, Wu MS (2016) Design and implementation of a GPON-based virtual open flow-enabled SDN switch. J Lightwave Technol 34(10):1–8 7. Lee J, Hwang I, Nikoukar A, Liem T (2013) Comprehensive performance assessment of bipartition upstream bandwidth assignment schemes in GPON. J Optic Commun Netw 5(11):1285–1295 8. Milanovic S (2014) Case study for a GPON deployment in the enterprise environment. J Netw 9(1):42–47 9. Ricciardi S, Santos-Boada G, Careglio D, Domingo-Pascual J (2012) GPON and EP2P: a techno-economic study. In: Proceedings of the 17th European conference on networks and optical communications (NOC), pp 1–6 10. Koonen AMJ (2006) Technologies and applications of FTTx. Published at 19th annual meeting of the IEEE, lasers and electro-optics society, LEOS 2006 11. Nesset D, Appathurai S, Davey R (2008) Extended reach GPON using high gain semiconductor optical amplifiers. In: Proceedings of the conference on optical fiber communication/national fiber optic engineers (OFC/NFOEC), pp 8–16

64 Md. H. Ali and M. H. Ali 12. Rodrigues C, Gamelas A, Carvalho F, Cartaxo A (2011) Evolution of FTTH networks based on radio-over-fibre. In: Proceedings of the 13th international conference on transparent optical networks, pp 7–12 13. Guimarães MA, Rocha LD (2011) Fractional E1 transport in gigabit passive optical network. In: Proceedings of the SBMO/IEEE MTT-S international microwave and optoelectronics conference (IMOC 2011), pp 44–50 14. Chu A, Poon KF, Ouali A (2013) Using ant colony optimization to design GPON-FTTH networks with aggregating equipment. In: Proceedings of the IEEE symposium on computa- tional intelligence for communication systems and networks (CIComms), pp 1–6 15. Róka R (2015) Analysis of possible exploitation for long reach passive optical networks. In: Proceedings of the 4th international conference on simulation and modeling methodologies, technologies and applications (SIMULTECH), pp 123–130 16. OptSIM user manual, Available online: https://www.lmd.polytechnique.fr

Chapter 6 A Priority-Based Deficit Weighted Round Robin Queuing for Dynamic Bandwidth Allocation Algorithm in Gigabit Passive Optical Network Md. Hayder Ali and Mohammad Hanif Ali 1 Introduction Accessing Internet is going to be a fundamental right like other basic human rights. It is not a matter that how often a user uses or surf Internet, waiting for a web page to load is certainly an irritation. To overwhelm this condition, Internet surfing speed has enlarged meaningfully in the earlier era to retain step with the petition of end- users, innovative amenities, and bandwidth-hungry claims. These anxieties embrace as hypermedia content-based e-commerce, video on claim, high classification TV, IPTV, online gaming, social media, etc. The communication protocols for hyper- media circulation have conventional a great transaction of attention in the earlier few ages. Since multimedia traffic must provision numerous types of traffic simulta- neously, it is crucial to process data according to its characteristics. Thus, protocol originators have to grasp the features of traffic and select a processing method suit- able for the performance requirements. For instance, actual acoustic traffic in a voice service requires rapid transmission, but the loss of a small amount of audio infor- mation is tolerable. On the opposite site, the transfer of a text file should guar- antee 100% reliable transfer; real-time delivery is not of primary importance in this case. Real-time video service, such as video on demand (VOD), requires not only rapid transfer but also high reliability. When a piece of video information is lost, its quality of service (QoS) is degraded. Therefore, multimedia communication proto- cols should be premeditated to afford the performance requirements of a wide range of multimedia services [1–7]. Md. H. Ali (B) · M. H. Ali 65 Jahangirnagar University, Dhaka, Bangladesh e-mail: [email protected] M. H. Ali e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2021 M. Shorif Uddin et al. (eds.), Intelligent Energy Management Technologies, Algorithms for Intelligent Systems, https://doi.org/10.1007/978-981-15-8820-4_6

66 Md. H. Ali and M. H. Ali The other parts of this paper are organized as follows. Proposed DBA poling mechanism for PMDWRR is introduced in Sect. 2. In Sect. 3, Pseudo code for PMDWRR is briefly described. Simulation scenario is in Sect. 4 and performance analysis between WRR, DWRR, and PMDWRR and stated in Sect. 5. Finally, Sect. 6 draws a conclusion to this paper. 2 Proposed DBA Poling Mechanism Design for Modified Deficit Weighted Round Robin 1. Start 2. ONUs/ONTs calculate the amount of received data 3. ONUs/ONTs predict the size of data in waiting time 4. ONUs/ONTs calculate the traffic for T-CONTs ( T-CONT1 to T-CONT4) 5. ONUs/ONTs enhance the prediction of accuracy by calculating arriving in waiting time (PTi,n) for n-th Cycle. 6. OLT calculates the arithmetic mean for the data received in waiting time. 7. Repeat step 2 to 6 for n-th cycle. 8. Then 9. Calculate the difference for the size of data received in waiting time at ONTi/ONUi in the n-the cycle. 10. Calculate the size of traffic prediction arriving in waiting time at ONTi/ONUi in the n-th cycle by adding a weighting factor. 11. Increase the average order to ensure the prediction accuracy. 12. OLT grant frame size or allocate the frame size for priority traffic. 13. ONTs/ONUs send data according to OLT’s priority grant frame. 14. End To execute the above algorithm for GPON networks, it is necessary to maintain a queuing system. DWRR algorithm is suitable for this. As it provides protection among different flows, it overcomes the limitations of strict PQ (Fig. 1). 3 Pseudo Code for Priority-Based Deficit Weighted Round Robin The pseudo code in this section does not describe the procedure of any exact DWRR execution. Although each execution will vary from this model, studying the samples and outlining the pseudo code will make it calmer to recognize the explicit enterprise results that are compulsory to make for executions. The array flexible Deficit_Counter is reset to zero. In this sample, the queues are numbered 1 to n, where n is the extreme number of lines on the output port:

6 A Priority-Based Deficit Weighted Round Robin … 67 OLT ONUi/ONTi t1 t2WaiƟng Time t3 REPORT Message GATE Message An Cycle Fig. 1 DBA polling mechanism diagram for PMDWRR FOR i = 1 to n /* Visit each queue index */ Deficit_Counter[i] = 0 /* Initialize Deficit_Counter[i] to 0 */ ENDFOR i = the index of the queue that will grip the new package IF (ExistsInActiveList(i) = FALSE) THEN /*IF i not in ActiveList */ InsertActiveList(i) /* Add i to the end of ActiveList */ Deficit_Counter[i] = 0 /*Reset queue Deficit_Counter[i] to 0*/ ENDFOR Enqueue packet to Queue[i] /* Place packet at finish of queue i */ END Enqueue // Removes the element with the highest priority form the list void pop (Node** head) { Node* temp = *head; (*head) = (*head)->next; free(temp); } // Function to push according to priority void push (Node** head, int d, int p) { Node* start = (*head); Whenever an index is at the head of the ActiveList, the function Dequeue () transmits up to Deficit_Counter[i] + Quantum[i] worth of bytes from queue.

68 Md. H. Ali and M. H. Ali Dequeue () While (TRUE) DO IF (ActiveList is NotEmpty) THEN i = the index at the head of the ActiveList Deficit_Counter[i] = Deficit_Counter[i] + Quantum[i] WHILE (Deficit_Counter[i] > 0 AND NOT Empty (Queue[i])) DO PacketSize = Size (Head (Queue[i])) IF (PacketSize <= Deficit_Counter[i]) THEN Transmit packet at head of Queue[i] Deficit_Counter[i] = Deficit_Counter[i] – PacketSize /* To find out the largest traffic volume */ /*Declare largest as integer*/ Set largest to 0 FOR EACH value in A DO IF A[n] is greater than largest THEN largest A[n] ENDIF END FOR Dequeue the largest first ELSE Break /*exit this while loop*/ ENDIF ENDWHILE IF (Empty (Queue[i])) THEN Deficit_Counter[i] = 0 RemoveFromActiveList(i) ELSE InsertActiveList(i) ENDIF ENDIF ENDWHILE END Dequeue 4 Simulation Design The simulation is done by using OptSIM (RSoft System Suit, version-2016.06) Simu- lation software. The simulation scenario is like bellow Fig. 2. It is composed of one OLT (Optical Line Terminal), one 1:32 Splitter and then thirty-two ONUs/ONTs. Each ONU/ONT has four signal sources producing T-CONT1 to T-CONT4 services respectively. Before each ONT, there is Bandwidth Analyzer to measure the consumed bandwidth of individual ONU/ONT. The OLT’s up and down link rates are 1.25 and 2.48 Gbps. Each ONU/ONT is 20 km far from the OLT.

6 A Priority-Based Deficit Weighted Round Robin … SpliƩer 69 Fig. 2 Simulation scenario in OptSIM simulator BA BA ONT-1 OLT BA ONT-2 ONT-3 BA ONT-32 5 Result Analysis The data has captured from OptSIM simulator and analyzed by MATLAB. Delay measurement and usage bandwidth are calculated by MATLAB coding. Figure 3 stated that number of end-users is more in PMDWRR. Considering 2GBPS bandwidth allocation, DWRR is allowing 17 users, MDWRR is allowing 20 users, while PMDWRR is allowing 23 users. Bandwidth is shared with maximum users which satisfy the properties of GPON system. Figure 4 shown that delay per user is minimum in PMDWRR. Considering 25 users in the system, DWRR shows 3 ms delay, MDWRR shows 2.4 ms delay while Fig. 3 Bandwidth allocation versus number of users

70 Md. H. Ali and M. H. Ali Fig. 4 Delay versus number of users PMDWRR shows only 1.7 ms/user. The per user delay is less in PMDWRR and it increases the system performance. 6 Conclusion In this paper, have partially modified to set priority into the existing MDWRR. In PMDWRR, ONU/ONT predict the data size and send allocation request to OLT, OLT increases the prediction accuracy. It could calculate the size of the queue and store the data size. By calculating the queue size, it can predict the priority traffic. It transmits the priority of the traffic and this way it could minimize the transmission delay and it satisfies the full service of QoS requirements of GPON system. References 1. Tae Il J, Jae Ho J, Sung Jo K (1997) An efficient scheduling mechanism using multiple thresholds for multimedia traffic in ATM switching nodes. In: Proceedings of the 22nd IEEE conference on local computer networks (LCN’97) 2. Michael P Mc, Martin M, Martin R (2004) Ethernet PONs: a survey of dynamic bandwidth allocation (DBA) algorithms. IEEE Commun Mag 42(8), 1–15 3. Taeck-Geun K, Sook-Hyang L, June-Kyung R (1998) Scheduling algorithm for real-time burst traffic using dynamic weighted round robin. In: Proceedings of the IEEE international symposium on circuits and systems (Cat. No.98CH36187), pp 506–509 4. Patel Z, Dalal U (2014) Design and implementation of low latency weighted round robin (LLWRR) scheduling for high speed networks. Int J Wirel Mob Netw (IJWMN) 6(4):59–71 5. Ji-Young K, Ji-Seung N, Doo-Hyun K (2002) A modified dynamic weighted round robin cell scheduling algorithm. ETRI J 24(5):360–372

6 A Priority-Based Deficit Weighted Round Robin … 71 6. Ouni R, Bhar J, Torki K (2013) A new scheduling protocol design based on deficit weighted round robin for QoS support in ip networks. J Circ Syst Comput 22(3):1–21 7. Lenzini L, Mingozzi E, Stea G (2006) Bandwidth and latency analysis of modified deficit round robin scheduling algorithms. In: Proceedings of the 1st international conference on performance evaluation methodologies and tools (VALUETOOLS 2006)

Chapter 7 Protection of Six-Phase Transmission Line Using Demeyer Wavelet Transform Gaurav Kapoor 1 Introduction An increase in the inevitability of electrical power has been perceived by the people of the modern generation. The electrical power transfer potentiality of the currently operating power transmission systems ought to be augmented in order to assist the significant increase in the necessity of electrical energy. Thus, accurate recognition of the faults in the SPTL turns out to be very decisive for mitigating the loss of gain and providing fast renovates. Numerous newly reported research works addressed the issue of FD and FC in TL’s. The discussion of few researches is shown in brief here in this section. In [1], FSM-based technique has been used for TL protection. In [2, 4, 7], WT and MM have been applied for the protection of SPTL. VMD has been employed for disturbance recognition in power system [3]. POVMD and WPNRVFLN have been used for FD and FPR in SCDCTL [5]. DM and MM have been used for high impedance fault recognition in TL [6]. Alienation-based technique has been proposed for TTTL protection [8]. In [9], the theory of TW has been reported and the combination of VMD and TEO has been applied for HVDC TL protection. VMD and HT have been used for micro-grid protection [10]. In this work, the demeyer wavelet transform (DMWT) has been used for the protection of six-phase transmission line (SPTL). No such type of work has been reported yet to the best of the knowledge of author. The results exemplify that the DMWT efficiently recognizes and categorizes the faults and the consistency of the DMWT is not affected by variation in the fault factors of SPTL. G. Kapoor (B) 73 Modi Institute of Technology, Kota, India e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2021 M. Shorif Uddin et al. (eds.), Intelligent Energy Management Technologies, Algorithms for Intelligent Systems, https://doi.org/10.1007/978-981-15-8820-4_7

74 Six Phase G. Kapoor Transmission 138 kV source Load-1 Line 138 kV source Relay Bus-1 Fault Load-2 Fig. 1 The schematic of SPTL Bus-2 This paper is structured as follows: The specifications of SPTL are presented in Sect. 2. The flow-diagram for DMWT is presented in Sect. 3. Section 4 is dedicated to the discussion of results. Section 5 concludes the paper. 2 The Specifications of SPTL Figure 1 shows the schematic of SPTL. The schematic consists of 138 kV, 60 Hz SPTL of 68 km, connected to a 138 kV voltage source at one end and loads at the other end. The SPTL is divided into two parts of length 34 km each. The CT’s and relay are connected at bus-1 to protect the full length of SPTL. 3 The Flow Chart of DMWT Figure 2 illustrates the process for the DMWT. The steps are shown beneath. Step 1 Simulate the SPTL for creating faults and produce the post-fault currents. Step 2 Use DMWT to examine the post-fault currents for characteristics retrieval and determine the range of DMWT output. Step 3 The phase will be proclaimed as the faulted phase if its DMWT output has a larger amplitude as compared to the healthy phase in a faulty situation. 4 Performance Assessment The simulations have been performed for the near-in relay faults, far-end relay faults, converting faults, inter-circuit faults, and multi-position faults with the objective of verifying the feasibility of the DMWT. In the separate subcategories, the outcomes of the work are investigated.

7 Protection of Six-Phase Transmission Line … 75 Record six phase current Signal processing using DMWT Evaluate DMWT coefficients Is |DMWT coefficient| faulted phase No > No fault |DMWT coefficient| healthy phase Yes Fault detection and Faulty phase recognition Fig. 2 Flow diagram showing DMWT process 4.1 The Efficacy of DMWT for Healthy Condition Figure 3 shows the SPTL currents and voltages for no-fault. The DMWT outputs for no-fault are shown in Fig. 4. Table 1 reports the results of DMWT for healthy situation. Fig. 3 Six-phase current and voltage waveforms for no-fault

76 G. Kapoor abc def Fig. 4 DMWT outputs of six-phase currents for no-fault Table 1 Results of DMWT for no-fault DMWT output Phase-A Phase-B Phase-C Phase-D Phase-E Phase-F 154.5936 133.9446 188.1193 175.9988 152.7358 155.3587 4.2 The Efficacy of DMWT for Near-in Relay Faults The efficiency of the DMWT is investigated for the near-in relay faults on the SPTL. Figure 5 exemplifies the SPTL currents for ABDFG near-in relay fault at 5 km at 0.0535 s among RF = 2 and RG = 4 . Figure 6 illustrates the DMWT output for ABDFG fault. The fault factors for the fault cases are: T = 0.0535 s, RF = 2 , and Fig. 5 ABDFG near-in relay fault at 5 km at 0.0535 s among RF = 2 and RG = 4

7 Protection of Six-Phase Transmission Line … 77 ab c de f Fig. 6 DMWT output for ABDFG near-in relay fault at 5 km at 0.0535 s Table 2 Results of DMWT for near-in relay faults Fault type DMWT output Phase-A Phase-B Phase-C Phase-D Phase-E Phase-F 905.7343 62.6620 1.4003 × ABDFG 975.5428 1.5645 × 82.1565 103 (5 km) 103 86.5100 44.6221 50.1690 ABCG 1.1302 × 1.0896 × 1.2082 × 151.0041 1.3309 × 1.0599 × (6 km) 103 103 103 103 103 1.1865 × 993.3149 208.7100 BCEFG 208.6318 1.1057 × 1.0163 × 103 (7 km) 103 103 1.9163 × 1.8024 × 1.1828 × 103 103 103 ABCDEG 1.4257 × 1.1870 × 1.1547 × (8 km) 103 103 103 DEFG 53.2673 46.4882 60.4460 (9 km) RG = 4 . Table 2 details the results of the DMWT for near-in relay faults. It is seen from Table 2 that the DMWT precisely detects the near-in relay faults. 4.3 The Efficacy of DMWT for Far-End Relay Faults The DMWT has been explored for the far-end relay faults. Figure 7 illustrates the SPTL currents for ABCEFG far-end relay fault at 63 km at 0.1 s among RF = 1.15 and RG = 2.15 . Figure 8 shows the DMWT output for ABCEFG fault. Table 3 reports the results for far-end relay faults. It is inspected from Table 3 that the effectiveness of the DMWT remains unaffected for the far-end relay faults.

78 G. Kapoor Fig. 7 ABCEFG far-end fault at 63 km at 0.1 s among RF = 1.15 and RG = 2.15 a bc def Fig. 8 DMWT output for ABCEFG far-end relay fault at 63 km at 0.1 s Table 3 Results of DMWT for far-end relay faults Fault type DMWT output Phase-A Phase-B Phase-C Phase-D Phase-E Phase-F 158.1244 2.5117 × ABCEFG 3.8060 × 7.9079 × 4.0745 × 3.4767 × 103 165.9761 103 197.3903 (63 km) 103 103 103 157.4620 5.1810 × 4.8779 × ABCG 4.8194 × 8.5085 × 4.3048 × 103 2.7625 × 103 (64 km) 103 103 103 5.8480 × 103 153.9397 103 4.5843 × DEFG 142.1903 137.1558 124.3455 3.1028 × 103 211.9889 (65 km) 103 197.5127 ABDEG 5.8668 × 6.2243 × 166.0415 (66 km) 103 103 ABCDG 3.0659 × 6.4599 × 3.7578 × (67 km) 103 103 103

7 Protection of Six-Phase Transmission Line … 79 Fig. 9 SPTL currents for multi-position ABG fault at 42 km and DEFG fault at 26 km abc def Fig. 10 DMWT output for multi-position ABG fault and DEFG fault 4.4 The Efficacy of DMWT for Multi-position Faults The DMWT is tested for different cases of multi-position faults. Figure 9 depicts the currents when the SPTL is simulated for the multi-position ABG fault at 42 km and DEFG fault at 26 km at 0.0725 s among RF = 1.5 and RG = 1.25 . Figure 10 shows the DMWT output for ABG and DEFG multi-position fault. Table 4 presents the results for different position-position faults. 4.5 The Efficacy of DMWT for Inter-circuit Faults The DMWT is tested for different cases of inter-circuit faults. Figure 11 depicts the currents when the SPTL is simulated for the inter-circuit ABCG and FG fault at 30 km at 0.1325 s among RF = 2.75 and RG = 1.85 . Figure 12 depicts the DMWT output for ABCG and FG inter-circuit fault. Table 5 reports the results for

80 G. Kapoor Table 4 Results of DMWT for multi-position faults FT-1 FT-2 DMWT output Phase-C Phase-D Phase-E Phase-F (km) (km) Phase-A Phase-B 6.5372 × 2.9300 × 6.4875 × ABG DFFG 4.1866 × 4.4410 × 203.0695 103 103 103 179.0262 (42) (26) 103 103 2.4488 × 2.7339 × 103 2.7835 × 103 CG DFG 248.1627 247.6852 1.4210 × 103 4.1065 × 2.8471 × 4.5857 × (20) (48) 103 103 103 103 97.7596 DEFG BG 93.2050 2.7384 × 114.9325 132.6308 4.2729 × (35) (33) 103 103 2.6415 × ACG EFG 2.4514 × 111.0308 2.0142 × 103 163.331 (44) (24) 103 103 DG ABCG 4.8020 × 6.5308 × 2.4552 × (38) (30) 103 103 103 Fig. 11 SPTL currents for ABCG and FG inter-circuit fault at 30 km at 0.1325 s a bc def Fig. 12 DMWT output for ABCG and FG inter-circuit fault at 30 km at 0.1325 s

7 Protection of Six-Phase Transmission Line … 81 Table 5 Results of DMWT for inter-circuit faults Fault-1 Fault-2 DMWT output Phase-A Phase-B Phase-C Phase-D Phase-E Phase-F 159.5320 95.4246 3.3819 × ABCG FG 5.4287 × 5.0592 × 3.2656 × 103 103 103 103 3.2120 × 244.3075 245.0540 103 BG DG 264.7818 4.0431 × 260.5588 4.8352 × 3.5412 × 131.1642 103 103 103 201.4383 2.6028 × 4.5607 × ACG DEG 3.7037 × 176.3243 2.6696 × 103 103 103 103 1.0169 × 8.9809 × 4.5267 × 103 103 103 ABG EFG 5.7039 × 5.8627 × 195.1929 103 103 CG DEFG 271.4223 202.1157 3.9056 × 103 the inter-circuit faults. It is examined from Table 5 that the DMWT performs well for the recognition of inter-circuit faults. 4.6 The Efficacy of DMWT for Converting Faults The DMWT has been investigated for the converting faults. Figure 13 exemplifies the currents of the SPTL when initially the ABG fault at 35 km at 0.05 s is converted into the DEFG fault at 35 km at 0.15 s among RF = 2.50 and RG = 1.50 . Figure 14 exemplifies the DMWT output for the same fault case. The fault factors preferred for the additional fault cases are: FL = 35 km, RF = 2.50 and RG = 1.50 . Table 6 reports the results for different converting faults. Fig. 13 SPTL current when ABG fault at 0.05 s is converted into DEFG fault at 0.15 s at 35 km among RF = 2.50 and RG = 1.50

82 G. Kapoor abc def Fig. 14 DMWT output when ABG fault is converted into DEFG fault Table 6 Results of DMWT for converting faults Fault Converted DMWT output Phase-D Phase-E Phase-F type fault Phase-A Phase-B Phase-C 6.2976 × 7.2091 × 3.1614 × 103 ABG DEFG 816.6582 1.2847 × 196.2113 103 103 4.2094 × (0.05) (0.15) 103 5.6853 × 103 103 1.7459 × 173.1067 DG DEFG 159.9643 136.8911 198.3681 197.2483 103 (0.1) (0.2) 110.2217 3.2852 × 1.3585 × ABCG EG 1.5407 × 2.1744 × 1.6289 × 103 103 912.3268 (0.07) (0.16) 103 103 103 645.9092 2.0547 × ACG DEG 2.0178 × 171.0564 1.1652 × 103 (0.05) (0.2) 103 103 1.4708 × EFG ABG 5.8620 × 6.6895 × 208.7799 103 (0.1) (0.15) 103 103 5 Conclusion The demeyer wavelet transform (DMWT) is seemed to be very efficient under varied fault categories for the SPTL. The DMWT output of fault currents of the SPTL is assessed. The fault factors of the SPTL are varied and it is discovered that the variation in fault factors do not influence the fidelity of the DMWT. The outcomes substantiate that the DMWT has the competence to protect the SPTL beside different fault categories.

7 Protection of Six-Phase Transmission Line … 83 References 1. Yadav A, Swetapadma A (2016) A finite-state machine based approach for fault detection and classification in transmission lines. Elect Power Component Syst (Taylor and Francis) 44(1):43–59 2. Kapoor G (2018) Six phase transmission line boundary protection using wavelet transform. In: Proceedings of the 8th IEEE India international conference on power electronics (IICPE), Jaipur, India (2018) 3. Jena MK, Samantaray SR, Panigrahi BK (2017) Variational mode decomposition-based power system disturbance assessment to enhance WA situational awareness and post-mortem Analysis. IET Gener Transm Distrib 11(13):3287–3298 4. Kapoor G (2018) Fault detection of phase to phase fault in series capacitor compensated six phase transmission line using wavelet transform. Jordan J Elect Eng 4(3):151–164 5. Sahani M, Dash PK (2019) Fault location estimation for series-compensated double-circuit transmission line using parameter optimized variational mode decomposition and weighted P-norm random vector functional link network. Appl Soft Comput J (Elsevier), 1–18 (2019) 6. Sekar K, Mohanty MK (2018) Data mining-based high impedance fault detection using mathematical morphology. Comput Elect Eng (Elsevier) 69:129–141 7. Kapoor G (2018) Six phase transmission line boundary protection using mathematical morphology. In: Proceedings of the IEEE international conference on computing, power and communication technologies (GUCON), pp 857–861, Greater Noida, India (2018) 8. Rathore B, Shaik AG (2018) Fault analysis using alienation technique for three-terminal trans- mission line. In: Proceedings of the IEEE 2nd international conference on power, energy and environment: towards smart technology (ICEPE), pp 1–6, Shillong, India (2018) 9. Wang L, Liu H, Dai LV, Liu Y (2018) Novel method for identifying fault location of mixed lines. Energies 11(1529):1–19 10. Chaitanya BK, Yadav A, Pazoki M (2019) An improved differential protection scheme for micro-grid using time frequency transform. Elect Power Energy Syste (Elsevier) 111:132–143

Chapter 8 Analysis of Voltage Sag and Swell Problems Using Fuzzy Logic for Power Quality Progress in Reliable Power System Ankit Tandon and Amit Singhal 1 Introduction The electricity generation from conventional and non-conventional sources, its transmission from DC or AC system, its distribution to domestic and commercial consumers, and its utilization from human being and for industrial requirements makes a power system network. In a smart power system, the difficulties during this complete process are solved in a smart manner. Such type of disturbances includes mainly VAR controlling and power quality issues. Managerial controlling from load dispatch centers makes a power system better. Additionally, the fuzzy-based expert systems can be implemented so as to make the existing system smarter with higher accuracy. The problems due to power quality and problems arising are penalty of more utilization of solid-state switching devices, nonlinear load, electronic load, and switching load. The arrival of high rating semiconductor switches in distribution and transfer leaves current to be non-sinusoidal [1]. Electronic load causes voltage distortion and harmonic distortion. Power quality problems create onset of systems, sensitive equipment, data loss in the commuter, and MAL functioning in the memory, like computers, programmable logic controller, the protective and relay apparatus [1]. A voltages sag or voltage dip is a short duration decrease in rms voltage. It is caused by short circuit phenomena, overload, and starting of electric motors. If the value of RMS voltage reduces from 10 to 90 percent of the nominal voltage for a cycle to a minute duration, then phenomena of voltage sag is produced. Voltage sag is termed as sustained sag if this low voltage remains in the system for a longer duration up to a minute. Voltage swell is the contradictory of voltage sag. It is the phenomena of an increase in the level of voltage which occurs due to turn off of a A. Tandon (B) · A. Singhal 85 Department of Electrical Engineering, JIET, Jodhpur, India e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2021 M. Shorif Uddin et al. (eds.), Intelligent Energy Management Technologies, Algorithms for Intelligent Systems, https://doi.org/10.1007/978-981-15-8820-4_8

86 A. Tandon and A. Singhal heavy load in an electrical power system. Similar to sag, swell happens for a very short duration, and if it is of longer duration, then termed as sustained swell. Voltage sag mainly increases the problems of power quality. It affects the electricity distribution system and various industries which causes very high loss of power. Short duration voltage dips can cause the whole industry out from their normal operating condition. Generally, it is considered that the voltage dip as a source 10–90%. The main reasons include short circuit condition and breakdown voltage dips, lightning phenomena, and power surges. In overhead transmission and distri- bution systems, lightning phenomena is the main reason for producing voltage sag, with an approximate occurrence of 50% of total cases [2]. The following pie chart shows the main reasons for poor power quality and their approximate percentage. To minimize conflict of power quality in electrical devices, some well-organized detection and categorization techniques are necessary in the electrical power systems. Types of power quality disturbances which are based on visual check of waveforms by human being operators is a time consuming and laborious task. Also, it’s not always necessary that we can remove some important information from simple visual inspection [2]. The categorization of Power quality disturbances in an electrical power system has now become a very important job for proper development and design of preventive and corrective actions. In the mid of 1950s, artificial intelligence has emerged as a new stream of computer science engineering. There are many powerful tools produced by artificial intelligence, which are used practically in different engineering streams so as to resolve those complex engineering problems which require human intelligence. In this paper, fuzzy logic tool is used for the analysis of disturbances like voltage sag and swell [2, 3]. Fuzzy Controller Fuzzy controllers are very simple and it is a new addition to control theory. They consist of an input stage, a processing stage, and an output stage. Its design philosophy differs from all previous methods by containing dexterous knowledge in controller design. These FLCs are attractive choices where specific mathematical modeling formulas are not possible. It has good control robustness as compared to tradi- tional control scheme. It modifies a dexterous knowledge-based control strategy into automatic control strategy in essence [3].

8 Analysis of Voltage Sag and Swell Problems … 87 Knowledge Base Rules Crisp Input Fuzzifier Defuzzifier Crisp Output x y Fuzzy input Inference Fuzzy output sets sets Fig. 1 Architecture of a Fuzzy Controller A Mamdani-based fuzzy controller is used with maximum-minimum interference method. It performs the function in the following 4 stages which is shown in the following Fig. 1. (1) Fuzzification process (2) Rule base (3) Inference (4) Defuzzification process. Fuzzification is the procedure of changing real scalar value into a fuzzy value which is achieved with the different types of fuzzifiers. (Membership functions). Rule base scheme is used as a way to store and manipulate knowledge to understand the information in a useful manner. They are used in AI. FIS is a system that uses the theory of fuzzy set for mapping of inputs. The input is features while the output is classes of fuzzy classification. Defuzzification is the process of getting a scientific result in the Crisp logic, available fuzzy sets, and related membership degrees. So, the process of mapping of fuzzy to crisp set is called as Defuzzification. The planned fuzzy-based expert system is developed to classify short voltage disturbances which can be defined as instantaneous and temporary sag, swell, and interruption, which are shown in Fig. 2 [3]. In the study, the disturbance data are obtained from PQ monitoring in which the monitoring software by default has three different sampling frequencies of 0.4 kHz (128 cycle), 1.6 kHz (32 cycle), and 6.4 kHz (8 cycles) and each frame has 1024 samples. To process the raw disturbance data so as to remove features of the various distur- bances, preprocessing of the disturbance signals is required. Initially, fast Fourier transform analysis is used to separate the 8, 32, and 128 cycle waveforms. Then root mean square (rms) method is applied by first approximating the fundamental frequency profile of actual voltage waveform and determining the maximum and minimum voltage magnitudes. An advantage of this method is its simplicity, fast calculation, and less requirement of memory because rms voltage can be stored periodically instead of per sample [4].

88 A. Tandon and A. Singhal Power Quality Actual Data 8 Cycles FFT 128 Cycles Obtaining rms voltage 32 Cycles Obtaining rms voltage & & Obtaining rms voltage Extract fuzzy inputs & Extract fuzzy inputs Extract fuzzy inputs Fuzzy logic to identifying & classifying disturbances Instan. Sag Momen. Sag Instan. Momen. Interruption Interruption Non- Swell Non-Sag Instan. Swell Momen. Swell Non-Interruption Fig. 2 Design of the proposed fuzzy-expert system 2 Fuzzy Logic Inputs and Outputs The Mamdani-type fuzzy system with five inputs and three outputs has been considered in the proposed fuzzy-expert system. The five inputs include maximum voltage in p.u (Max-V), sag duration in second (Sag Durat), swell duration in second (Swell Durat), transient duration in second (Tran Durat), and minimum voltage in p.u (Min-V) [5]. The maximum voltage has been chosen as a fuzzy input variable so as to classify instantaneous and momentary swell whereas the minimum voltage is for differentiating between interruption and voltage sag. The sag duration is used as an input variable for classifying instantaneous and momentary sag whereas the swell duration is used to classify swell disturbance to instantaneous and momentary swell. The transient duration is chosen as a fuzzy input for distinguishing between instantaneous swell and impulsive transient disturbance [5]. The three FL outputs are Output1, Output2, and Output3 in which Output1 is designated for classifying instantaneous sag, non-sag, and momentary sag, Output2 for classifying instantaneous swell, non-swell, and momentary swell, and Output3 for classifying instantaneous interruption, non-interruption, and momentary interruption [6].

8 Analysis of Voltage Sag and Swell Problems … 89 3 Membership Functions The input and output variables are represented by the most common shape of member- ship functions which are either in trapezoidal or triangular forms and bell curves are also used but the shape is less important than no. of curves. The range of input variables and thresholds are chosen in accordance with the respective disturbance definition as defined in the IEEE Std. 1159-1995. The following Tables 1 and 2 represent fuzzy sets for Input and Output Variables [7, 8]. The membership functions defined for the five input variables are as shown in Figs. 3, 4, 5, 6, and 7. The output variables are defined by three membership functions as shown in Figs. 8, 9 and 10. If-Then rules (30) have been generated for classifying sag, swell, and interruption disturbances. These rules are represented in the following form: IF premise THEN consequent. Table 1 Fuzzy sets defined for the input variables Membership Input 1: Input 2: Sag Input 3: Swell Input 4: Input 5: Transient Absolute Function Maximum duration (s) Duration (s) Duration (s) Minimum voltage (p.u.) voltage (p.u.) ESH Extremely VL 1. L ESH ESH short Very Low SH Low Extremely Extremely Short L SH Low short short Short – M 2. M VSH VSH Medium – Medium Very Short Very Short H High 3. H SH SH – High Short Short 4. VH M M Very High Medium Medium 5. EH – – Extremely High Table 2 Fuzzy sets defined Membership Output 1 Output 2 Output 3 for the output variables function 1 I sag I swell I interrupt Instantaneous Instantaneous Instantaneous 2 sag swell interruption N sag N swell 3 Non sag Non swell N interrupt Non M sag M swell interruption Momentary Momentary sag swell M interrupt Momentary interruption

90 A. Tandon and A. Singhal Fig. 3 Maximum voltage input membership function Fig. 4 Sag duration input membership function Fig. 5 Swell duration input membership functions

8 Analysis of Voltage Sag and Swell Problems … 91 Fig. 6 Transient duration input membership functions Fig. 7 Absolute minimum voltage input membership function Fig. 8 Output1 membership function

92 A. Tandon and A. Singhal Fig. 9 Output2 membership function Fig. 10 Output3 membership function Example of the generated rules for identifying sag, swell, and interruption and classifying them to instantaneous, momentary, and non-sag, swell and interruption are given as follows: (1) If (Max-V is L) and (Sag Durat is SH) and (Swell Durat is ESH) and (Tran Durat is ESH) and (Min-V is L) then (Output1 is I sag) (Output2 is N swell) (Output3 is N interrupt). etc. 4 Conclusion During the contingency analysis of an electrical power system, the variation in the level of voltage and current are the two important issues that affects the reliability a lot. In this paper, a fuzzy logic-based system is developed so as to analyze such types